Adaptive feedback method for cellular communication

1. An adaptive feedback method for a cellular communication system having a plurality of Mobile Stations (MSs) feeding back channel information to a Base Station (BS), comprising: selecting one of an opportunistic mode and a polling mode as a feedback mode between the BS and the plurality of MSs on the basis of system and traffic parameters; setting the BS and the plurality of MSs for the selected feedback mode; and feeding back the channel information from the plurality of MSs to the BS in the selected feedback mode, wherein the traffic parameters include a number ‘N’ of active MSs from among the plurality of MSs and the system parameters include a number of mini-slots ‘K’ carrying the channel information, and wherein the step of selecting the feedback mode includes: collecting the system and traffic parameters; computing a mode selection parameter from the system and traffic parameters; comparing the mode selection parameter with a predetermined threshold; and selecting the opportunistic mode if the mode selection parameter is greater than or equal to the threshold; and selecting the polling mode, if the mode selection parameter is less than the threshold.

2. The method of claim 1 , wherein the mode selecting parameter is obtained by Q=N−αK, where α is a tradeoff parameter for determining a switching point between the feedback modes.

3. The method of claim 2 , wherein the threshold is ‘0.’

4. The method of claim 1 , wherein the step of setting the BS and the plurality of MSs includes: optimizing control parameters associated with the selected feedback mode.

5. The method of claim 4 , wherein the control parameters includes a channel power threshold η and a random feedback probability p.

6. The method of claim 5 , wherein the step of optimizing the control parameters for the opportunistic mode includes: calculating a mean sum rate of all the plurality of MSs; and determining the channel power threshold η and the random feedback probability p to maximize the mean sum rate.

7. The method of claim 6 , wherein the mean sum rate is calculated according to: C ⁡ ( p , η ) = ∑ j = 1 N ⁢ ⁢ E ( χ j , S j , T ) ⁡ [ log 2 ⁡ ( 1 + ρ ⁢ ⁢ χ j ) ⁢ I ⁢ { S j = 1 } ] where χ j is the channel power of a j th MS, denoted by χ j =|h j | 2 S j is an event that the j th MS is selected at the BS, T is the number of the plurality of MSs whose channel powers are above the channel power threshold η, ρ is the average signal-to-noise ratio (SNR), and I{•} is an indicator function.

9. The method of claim 5 , wherein the step of feeding back the channel information includes: determining, at each of the plurality of MSs, whether a channel gain of the each of the plurality of MSs is greater than the channel power threshold η; and feeding back the channel information with the feedback probability p when the channel gain of the each of the plurality of MSs is greater than the channel power threshold η.

10. A Base Station (BS) of a cellular communication system including a plurality of BSs, comprising: a scheduler which schedules services for a plurality of Mobile Stations (MSs) and decides a service rate on the basis of channel information and traffic information with the respect to the services; and a central feedback controller which generates feedback control signals for controlling a selected feedback mode, wherein the BS operates in one of at least two feedback modes selected on the basis of a mode selection parameter and the plurality of MSs feed back channel information according to the selected feedback mode, wherein the mode selection parameter is selected on the basis of a number ‘N’ of active MSs from among the plurality of MSs and a number of mini-slots ‘K’ carrying the channel information, and wherein the central feedback controller includes: a mode selector which selects the feedback mode by comparing the mode selection parameter with a threshold; and an optimizer for optimizing the feedback mode selected by the mode selector and control parameters.

11. The BS of claim 10 , wherein the mode selection parameter is calculated according to Q=N−αK, where Q is the selection parameter, N is the number of the plurality of the MSs, K is the number of minislots, and α is a tradeoff parameter for deciding a mode switching point.

12. The BS of claim 11 , wherein the control parameters includes a channel power threshold η and a random feedback probability p.

13. The BS of claim 12 , wherein the optimizer optimizes the selected feedback mode for maximizing system throughput.

14. The BS of claim 13 , wherein the system throughput is a mean sum rate of the plurality of MSs.

15. The BS of claim 14 , wherein the optimizer selects the channel power threshold and the random feedback probability maximizing the mean sum rate as optimal control parameters.

16. The BS of claim 15 , wherein the mean sum rate is calculated according to: C ⁡ ( p , η ) = ∑ j = 1 N ⁢ E ( χ j , S j , T ) ⁡ [ log 2 ⁡ ( 1 + ρ ⁢ ⁢ χ j ) ⁢ I ⁢ { S j = 1 } ] where χ j is the channel power of a j th MS, denoted by χ j =|h j | 2 , S j is an event that the j th MS is selected at the BS, T is the number of the plurality of MSs whose channel powers are above the channel power threshold η, ρ is the average signal-to-noise ratio (SNR), and I{•} is an indicator function.